ICT Technical Seminar

Natural Cements for Repair and Renovation of Heritage Structures 27

th June 2013

Meeting at the Belgrave Square headquarters of the

Society of Chemical Industry, the Institute of

Concrete Technology held its summer seminar on 27

June, with the support of SCI and the Building Limes

Forum. Its title was Natural Cements for Repair and

Renovation of Heritage Structures, and it attracted a

diverse range of 39 participants, from Austria,

Germany, Ireland, the Netherlands, Poland,

Switzerland and the USA, as well as Great Britain.

Professor Peter Hewlett, President of the ICT,

introduced the theme of the day and made reference

to the SCI’s strapline, “where science meets

business”. He acknowledged that for concrete

specialists, dwelling on natural cements for render

repairs might seem “heresy”, but urged us, with a

twinkle in his eye, to “keep it friendly”. And so doing,

he warmly introduced the first speaker, Johannes

Weber, from the University of Applied Arts, Vienna.

Prof Weber expressed his pleasure to be back in the

“mother country” of cement and took a few minutes

to clarify definition and terminology. To set the

proceedings in context he described the EU-funded

projects RoCem and RoCARE. These were

international projects to explore, respectively, the

material properties and manufacturing process of

Roman cement, and the application of such materials

to the repair of historic building stock. Roman

cement was produced throughout Europe in the

Nineteenth Century and so was both varied and

widely applied. He described their common

properties and identified applications in masonry

mortars, mouldings, rendering and repairs. Cities

with a conspicuous legacy of Roman cement buildings

include Lviv, Prague, Budapest and Krakov. Perhaps

90% of pre 1920 buildings are rendered, and 25% of

these currently need repair. Far from being esoteric,

the RoCARE project addressed a pressing need.

Next Christophe Gosselin of Geotest SA, Switzerland,

turned to the subject of ‘Composition and Hydration

of some Roman (natural) Cements’. Dr Gosselin drew

attention to the high clay contents of the marlstones

used in Roman cement production, and the presence

of variable raw materials such as calcite, quartz, iron

sulphate. He also noted that the material was

dependant on calcination conditions and that there

was no standard of production, but also, that the

early age hydration of Roman cements is a very rapid

reaction leading to the formation of different

hydration products. His paper compared the

composition and hydration of different modern

Roman cements available on the market and

developed within the RoCem and RoCARE Projects.

The day’s programme allowed plenty of time for

discussion and the Chairman took questions for both

the opening speakers at this stage, before breaking

for coffee in the exhibition area.

After the break Dave Hughes, formerly Professor of

Construction Materials at Bradford, but now enjoying

the rewards of recent retirement, addressed the

topic, ‘Formulating mortars for use in restoration

practice’. He referred principally to a research

project that compared the performance of Roman

and Portland cement mortars, and those based on

limes, particularly in so far as they were used for

renders and cast elements. Tested in detail were

CEM 1, NHL 5 and three Roman cements (the latter

from Gartenau, Wietersdorfer and Vicat.) Prompted

by the small size of the squares of repair mortar used

at Durham cathedral – necessary as a consequence of

the very rapid setting of Roman cements from

Whitby – Dave’s project was led to the search for a

means of slowing the reaction time. Various forms of

retarder were tried, to no avail, so Dave turned to the

development of a deactivation process, now known

by the acronym ‘DARC’. DARC required the addition

of water to the sand, and storage for a preparatory

period. This affects the timing of the hydration

process and leads to a notable increase in workable

life, related to the quantity of water used and the

time stored. Yet the early hydration has little adverse

effect on the properties of the hardened mortar. The

commercial exploitation of this finding is now being

undertaken in Germany for the silo mortar market.

The addition of quicklime was also explored. The

results of testing their respective properties places

Roman cement between CEM 1 and NHL5. He

concluded that lime does not retard Roman cement,

though you might get a substandard product; 50%

addition is manageable, but it does reduce strength.

Turning from the laboratory to ‘Practical aspects of

restoring with mortars based on Roman Cements’

Jacek Olesiak of Remmers Polska rehearsed his

experience in the historic city of Krakow, where

Roman cement was widely used between 1870 and

1914. From private houses to palaces, schools to

banks, many buildings were rendered in this naturally

ochre material and left unpainted, embellished by

ornaments cast in Roman cement. However, in the

years since they have often been badly repaired and

inappropriately painted; Jacek pointed out some

examples of unwarranted decay. Conversely, the

Trade Academy of 1904-06 was restored in 2005-07

as the first full restoration to use one of the newly

developed RoCem materials. He described the

project. Super-heated water was used to remove

paint, and sandblasting to treat surfaces. Some

cracks and stains were exposed, but most of the

structure was adequate. A broken cornice was

repaired, using materials mixed on site and though

Portland cement was used to repair the ground floor,

as originally constructed, Roman cement was applied

to the first and upper floors. Ready-mixed dry

mortars were used in some instances, wetted during

the first two weeks. Where iron beams had corroded

and destroyed the Roman cement-based cover,

replacements were epoxy-coated. When the work

was completed some lime bloom occurred, but was

accepted by the client as a temporary phenomenon.

Jacek’s presentation was highly visual and prompted

considerable debate on the advisability of painting

and sealing against moisture ingress, whereas

questions for Dave Hughes included the reasons for

using citric acid as a retarder. (Answer: although old-

fashioned, it is cheap and recommended by Vicat.)

Animated discussion continued into the adjoining

Garden Room where there was an admirable lunch

and time to visit the accompanying exhibition.

Exhibitors included Cornish Lime, Fugro Aperio,

Lafarge Tarmac and the ICT itself.

After lunch, in a presentation entitled ‘Salt resistant

mortars’, Dr Barbara Lubelli addressed the issue of

the compatibility of original and repair materials. She

identified salt crystallization damage as one of the

most common causes of decay for bedding, pointing

and rendering mortar but observed that the

phenomenon had not been adequately explained.

Previous practice in resolving the problem had been

directed to increasing the strength of the mortar, the

replacement of lime with Portland cement, or on

reducing the moisture transport capacity by the

addition of silicone-based water repellents. These

solutions have often displayed limited resistance to

salt decay and a low compatibility with historical

buildings. Barbara’s research has started to explore

new possibilities for improving the durability of

mortars to salt damage; these include engineering of

the pore size, replacing of silicones with natural

organic water-repellent substances (such as linseed

oil) and mixing-in of inhibitors able to reduce the

harmfulness of salt crystallization (though this can

create efflorescence). Preliminary studies indicate

that these new directions show promise in obtaining

more durable mortars that are compatible with

historical buildings.

Paul Livesey confessed that, as a Portland cement

man, he had ‘seen the light’ some year ago in

Sunderland and persuaded Castle Cement to

introduced a wider range of lime-based products.

(The scene of his conversion – the Stadium of Light!)

He described a spectrum of binders, from agricultural

lime to hydraulic lime, natural cements to Portland,

and gave a chronological account of their

development towards modern materials. He focused

on the pioneering period from Smeaton (who tested

300 varieties of limestone in his search for a hydraulic

cement) through to the perfection of the Portland

cement production process and introduction of the

rotary kiln. Speaking authoritatively and elegantly,

he indicated the contributions of Semple and Higgins

in eighteenth century Britain, and Loriot and

Perronnet in France, before James Parker patented

Roman cement in the 1790s. The combination of clay

and limestone was researched extensively by Vicat,

while commercial production of natural and artificial

cements grow apace in Britain. Interestingly Paul

showed an historical photograph of the Roman

cement works at Whitby – the very same location

that had been illustrated earlier with a modern

picture by Dave Hughes. The account came to a

natural climax with Joseph Aspdin’s 1824 patent for

Portland cement and his son’s discovery of the

importance of clinkering in the 1840s.

Describing Scotland as geographically diverse, with

varied building styles in consequence, John Hughes

introduced Aberdeen as a city of granite – a low

porosity, high strength and durable material. Locally,

it seems, there have been problems with dampness

in nineteenth century buildings. Drainage failures are

a cause of water ingress, but in 30-40% of cases

dampness appears to be down to use of the wrong

pointing materials. To test this assumption, and

investigate what would make an appropriate re-

pointing and repair material, the University of the

West of Scotland won a couple of research projects

that John duly described. Work in 2008 looked at

calcium-loading in mortars, and found Portland

cement to be present in every mortar. The team

looked at hot mixes, including lump quicklime, and

various others including Portland cement mixes.

They tested for capillary arch and moisture

absorption, but lime works differently from Portland

cement, and laboratory testing has its limits. It is

more difficult to sample traditional mortars in situ.

Although most of the traditional quarries in Aberdeen

are no longer operating, Sclatti Quarry contains a

Council-owned store of stone from demolition

projects, to offset the shortfall of repair material.

From this source John and his team took 22 samples

of historic mortar – bedding (yellowish, crumbly) and

pointing (harder, greyish) – and subjected them to

testing. By 2011 the team had moved to developing

specifications, and were examining Strength,

Moisture and Bond – the most important factor.

There are still unresolved questions and Portland

cement is not consistent with Scottish conservation

practice, so Aberdeen Council is now looking at

natural hydrated lime. May be, John wondered,

Roman cement could be the answer?

Standing in for the scheduled speaker at very short

notice, event sponsor Phil Brown of Cornish Lime

described the application of Roman cement in

practice, drawing on the prime example of Hadlow

Tower, a grand folly that was rendered originally with

mortar based on Roman cement from Sheppey. The

restoration project got under way in 2009, using Vicat

materials, pigmented with stone dust as required.

Concentrating on the practical, Phil said that Roman

cement feels quite different on the trowel – you have

to use it to know – and that it is applied in three

coats: ‘fresh on fresh’. Reciting the craftsman’s

plaintive cri de coeur: “There’s never time to do it

right; but there is to do it twice”, Phil was delighted

to claim that the men at Hadlow had got it right first

time.

Then, leading into the open discussion session, there

was an impromptu addition to the programme. One

of the participants, Michael O’Reilly, was one such

craftsman and had come armed with a set of

photographs of his highly skilled work on recent

projects. A five-minute slide show followed, with

pictures of renovation work in Buckinghamshire and

at Slapton church in Cambridgeshire. Michael shared

with the audience his personal and very positive

views on the use of Roman cement for restoration.

Winding up the debate, Professor Hewlett

summarised the day’s proceedings, commenting on

themes such as the pros and cons of sealing Roman

cement render, and noting the material’s variability,

and expressing a pleasant surprise in the degree of

interest and expertise this niche sector had

generated in recent years.

Echoing these sentiments, the feedback from

speakers and audience alike has been generous:

“Congratulations on organising such a

brilliant conference last week. Excellent

speakers and everything ran to time, which

makes a pleasant change. It was the best

conference I have been to for a long time.”

[Alison Henry, English Heritage]

“A truly interesting and well organised

meeting!” [Prof Johannes Weber, Vienna];

and “… a perfect conference”

[Jacek Olesiak, Remmers Polska]

“It was an interesting mix of papers. Whilst

sad not to see Gerard there the substitute

paper from Vicat was inspirational.”

[Prof Dave Hughes, Bradford University]

“I would just like to say a big thank you for

organising yesterday’s conference as I

thoroughly enjoyed and found it most

informing. I think these topical one day

events where chemistry meets research, and

also site application, are fantastic.”

[Ed Hiam, Lime Green]

The full papers will be published in the 2013/14

edition of the Institute’s Yearbook.

Words: Edwin A.R. Trout, Executive Officer

Photos: Raman Mangabhai, Events Chairman

Institute of Concrete Technology